Information visualisation and data analysis using web mash-up systems

5.00
Hdl Handle:
http://hdl.handle.net/10547/584232
Title:
Information visualisation and data analysis using web mash-up systems
Authors:
Khan, Wajid
Abstract:
The arrival of E-commerce systems have contributed greatly to the economy and have played a vital role in collecting a huge amount of transactional data. It is becoming difficult day by day to analyse business and consumer behaviour with the production of such a colossal volume of data. Enterprise 2.0 has the ability to store and create an enormous amount of transactional data; the purpose for which data was collected could quite easily be disassociated as the essential information goes unnoticed in large and complex data sets. The information overflow is a major contributor to the dilemma. In the current environment, where hardware systems have the ability to store such large volumes of data and the software systems have the capability of substantial data production, data exploration problems are on the rise. The problem is not with the production or storage of data but with the effectiveness of the systems and techniques where essential information could be retrieved from complex data sets in a comprehensive and logical approach as the data questions are asked. Using the existing information retrieval systems and visualisation tools, the more specific questions are asked, the more definitive and unambiguous are the visualised results that could be attained, but when it comes to complex and large data sets there are no elementary or simple questions. Therefore a profound information visualisation model and system is required to analyse complex data sets through data analysis and information visualisation, to make it possible for the decision makers to identify the expected and discover the unexpected. In order to address complex data problems, a comprehensive and robust visualisation model and system is introduced. The visualisation model consists of four major layers, (i) acquisition and data analysis, (ii) data representation, (iii) user and computer interaction and (iv) results repositories. There are major contributions in all four layers but particularly in data acquisition and data representation. Multiple attribute and dimensional data visualisation techniques are identified in Enterprise 2.0 and Web 2.0 environment. Transactional tagging and linked data are unearthed which is a novel contribution in information visualisation. The visualisation model and system is first realised as a tangible software system, which is then validated through different and large types of data sets in three experiments. The first experiment is based on the large Royal Mail postcode data set. The second experiment is based on a large transactional data set in an enterprise environment while the same data set is processed in a non-enterprise environment. The system interaction facilitated through new mashup techniques enables users to interact more fluently with data and the representation layer. The results are exported into various reusable formats and retrieved for further comparison and analysis purposes. The information visualisation model introduced in this research is a compact process for any size and type of data set which is a major contribution in information visualisation and data analysis. Advanced data representation techniques are employed using various web mashup technologies. New visualisation techniques have emerged from the research such as transactional tagging visualisation and linked data visualisation. The information visualisation model and system is extremely useful in addressing complex data problems with strategies that are easy to interact with and integrate.
Citation:
Khan, W. (2014) 'Information Visualisation and Data Analysis Using Web Mash-up Systems ' PhD Thesis. University of Bedfordshire.
Publisher:
University of Bedfordshire
Issue Date:
Dec-2014
URI:
http://hdl.handle.net/10547/584232
Type:
Thesis or dissertation
Language:
en
Description:
A thesis submitted in partial fulfilment for the degree of Doctor of Philosophy
Appears in Collections:
PhD e-theses

Full metadata record

DC FieldValue Language
dc.contributor.authorKhan, Wajiden
dc.date.accessioned2015-12-21T11:08:16Zen
dc.date.available2015-12-21T11:08:16Zen
dc.date.issued2014-12en
dc.identifier.citationKhan, W. (2014) 'Information Visualisation and Data Analysis Using Web Mash-up Systems ' PhD Thesis. University of Bedfordshire.en
dc.identifier.urihttp://hdl.handle.net/10547/584232en
dc.descriptionA thesis submitted in partial fulfilment for the degree of Doctor of Philosophyen
dc.description.abstractThe arrival of E-commerce systems have contributed greatly to the economy and have played a vital role in collecting a huge amount of transactional data. It is becoming difficult day by day to analyse business and consumer behaviour with the production of such a colossal volume of data. Enterprise 2.0 has the ability to store and create an enormous amount of transactional data; the purpose for which data was collected could quite easily be disassociated as the essential information goes unnoticed in large and complex data sets. The information overflow is a major contributor to the dilemma. In the current environment, where hardware systems have the ability to store such large volumes of data and the software systems have the capability of substantial data production, data exploration problems are on the rise. The problem is not with the production or storage of data but with the effectiveness of the systems and techniques where essential information could be retrieved from complex data sets in a comprehensive and logical approach as the data questions are asked. Using the existing information retrieval systems and visualisation tools, the more specific questions are asked, the more definitive and unambiguous are the visualised results that could be attained, but when it comes to complex and large data sets there are no elementary or simple questions. Therefore a profound information visualisation model and system is required to analyse complex data sets through data analysis and information visualisation, to make it possible for the decision makers to identify the expected and discover the unexpected. In order to address complex data problems, a comprehensive and robust visualisation model and system is introduced. The visualisation model consists of four major layers, (i) acquisition and data analysis, (ii) data representation, (iii) user and computer interaction and (iv) results repositories. There are major contributions in all four layers but particularly in data acquisition and data representation. Multiple attribute and dimensional data visualisation techniques are identified in Enterprise 2.0 and Web 2.0 environment. Transactional tagging and linked data are unearthed which is a novel contribution in information visualisation. The visualisation model and system is first realised as a tangible software system, which is then validated through different and large types of data sets in three experiments. The first experiment is based on the large Royal Mail postcode data set. The second experiment is based on a large transactional data set in an enterprise environment while the same data set is processed in a non-enterprise environment. The system interaction facilitated through new mashup techniques enables users to interact more fluently with data and the representation layer. The results are exported into various reusable formats and retrieved for further comparison and analysis purposes. The information visualisation model introduced in this research is a compact process for any size and type of data set which is a major contribution in information visualisation and data analysis. Advanced data representation techniques are employed using various web mashup technologies. New visualisation techniques have emerged from the research such as transactional tagging visualisation and linked data visualisation. The information visualisation model and system is extremely useful in addressing complex data problems with strategies that are easy to interact with and integrate.en
dc.language.isoenen
dc.publisherUniversity of Bedfordshireen
dc.subjectinformation visualisationen
dc.subjectdata analysisen
dc.subjectweben
dc.subjectmash-up systemsen
dc.subjectmash-upsen
dc.subjecte-commerceen
dc.subjectG500 Information Systemsen
dc.titleInformation visualisation and data analysis using web mash-up systemsen
dc.typeThesis or dissertationen
dc.type.qualificationnamePhDen_GB
dc.type.qualificationlevelPhDen
dc.publisher.institutionUniversity of Bedfordshireen
This item is licensed under a Creative Commons License
Creative Commons
All Items in UOBREP are protected by copyright, with all rights reserved, unless otherwise indicated.